Air Quality Impact Assessment for Dorstfontein Coal Mine East Pit 1 Extension

Transcription

1 Air Quality Impact Assessment for Dorstfontein Coal Mine East Pit 1 Extension Report Prepared for Exxaro Coal Central (Pty) Ltd Report Number /AQIA Report Prepared by August 2017

2 SRK Consulting: : East Dorstfontein AQIA Page i Air Quality Impact Assessment for Dorstfontein East Coal Mine Pit 1 Extension Exxaro Coal Central (Pty) Ltd SRK Consulting (South Africa) (Pty) Ltd. Section A Second Floor Norfolk House 54 Norfolk Terrace Westville 3630 South Africa Durban@srk.co.za website: Tel: +27 (0) Fax:+27 (0) SRK Project Number August 2017 Compiled by: D. Naidoo (Pr. Sci. Nat.) Senior Scientist Peer Reviewed by: V.S. Reddy (Pr. Sci. Nat.) Partner dnaidoo@srk.co.za Authors: D. Naidoo and A. Murray-Rogers

3 SRK Consulting: : East Dorstfontein AQIA Page ii Executive Summary Introduction SRK Consulting (South Africa) (Pty) Ltd was appointed by Exxaro Coal Central (Pty) Ltd to undertake the necessary Environmental Impact Assessment in support of an application for an Environmental Authorisation for the expansion of its Dorstfontein East Mine in Mpumalanga. This report provides a baseline description of the air quality and meteorological conditions for the study area and assesses potential impacts associated with the proposed expansion may have on the air quality in the area of the mine. The outcomes of the impact assessment were used to identify management measures required to manage potential negative impacts. Background and Project Description Exxaro plans to undertake the necessary EA process for the extension of Pit 1, the pipeline from Dorstfontein West to Dorstfontein East. This supplementary footprint does not fall within the existing Environmental Management Plan (EMP), thus, authorization will be sought for the proposed infrastructure in terms of the National Environmental Management Act (Act No 107 of 1998) (NEMA) and National Water Act (Act No. 38 of 1996) (NWA). Dorstfontein East is an opencast mine located approximately 18 km from Ga-Nala, formerly known as Kriel. The mine is located within the Emalahleni Municipality and began operations in Currently, only opencast mining is practised and underground mining is an option that is currently being investigated. Two opencast pits are currently mined on the eastern and western sides of the resource area with mining taking place both in the 2 and 4 coal seams. The current activities have been approved by both the Department of Mineral Resources (DMR) and the Department of Water and Sanitation (DWS). Dorstfontein Coal Mines (Pty) Ltd is now planning to extend its operations on the western side of the mine referred to as Pit 1 Extension. The area is part of the current approved mining right area but it has not been included in the current Environmental Management Programme Report and Life of Mine. Impact Assessment Summary The impact is considered to be of low significance before and after management (Table ES 1). The impact is insignificant as the expected emissions of dust and expected contribution to ambient air quality during the construction phase for the proposed project with management will be below the NAAQS and NDCR. The impacts will be closer to construction activities and will decrease further away. Towns and surrounding farms in range of the construction activities will not be impacted as the expected contribution to levels of dust are expected to be low. Table ES 1: Construction phase impact rating Impact Consequence Likelihood Significance Severity Spatial Duration Frequency Frequency Rating of Activity of Impact Impact before management Impact after management Mitigation Rating Low Low

4 SRK Consulting: : East Dorstfontein AQIA Page iii For the construction phase the impact is considered to be of medium-high significance before management and of low significance after management (Table ES 2). The impact before management is significant because the expected emissions of dust and expected contribution to ambient air quality during the operational phase are above the 24 hour NAAQS for PM 10 and PM 2.5, however below the NDCR for dust pollutants. The impact after management measures are implemented is insignificant as the expected emissions of dust and expected contribution to ambient air quality during the operational phase for are low and below the 24 hour NAAQS and the NDCR for dust pollutants beyond the fenceline of the mine. The impacts are closer to operational activities and decrease further away. Towns and surrounding farms in range of the operational activities will not be impacted as the expected contribution to levels of dust is not harmful. Table ES 2: Operational phase impact assessment Impact Consequence Likelihood Significance Severity Spatial Duration Frequency Frequency Rating of Activity of Impact Impact before management Impact after management Mitigation Rating Mediumhigh Low Conclusions Based on the findings of this assessment the following were concluded: Based on the meteorological data from the on-site weather station, the climate in the project area is expected to be seasonal, with distinct warm months (September to February) and cool months (March to August). Rainfall is higher during the months from October to April and lower from May to September. The prevailing winds for this period are from the north, west-southwest and southeast, with lower occurrences from the east, east-southeast and south-southeast. The average wind speed measured for all hours is 4.47 m/s with maximum speeds less than 11.1 m/s and with calms (<0.5 m/s) of 1.72%. The monthly dust fallout concentrations shows that dust fallout concentration have exceeded the Residential Area limit of 600 mg/m 2 /day 361 times out of 1118 (32.3%) recorded dust fallout measurements. The Non-residential Area limit was exceeded 27 times out of 119 (22.7%) dust fallout measurements. The towns in the surrounding area i.e. Kriel, Tubelihle and Boskrans, have been identified as potential areas of impact, at the commencement of the study. The dispersion model was set up to predict PM 10, PM 2.5 and dust fallout concentrations from the proposed development at these towns and surrounding areas. Based on the nature of the activities and the dispersion modelling results, the receptors are unlikely to be impacted should management measures be implemented. With respect to dust (dust fallout, PM 10 and PM 2.5) concentrations, the predicted emissions resulting from the proposed operations were determined to be medium-high without management and low with management measures in place. Dust fallout concentrations will not exceed the NDCR limits at the closest sensitive receptors. Based on the findings of this assessment the proposed project will result in dust generation, however, the concentrations will be low and below the NAAQS (PM 10 and PM 2.5) and the NDCR (dust fallout) at the closest receptors, should management measures be implemented. Receptors in close proximity will not be impacted by the proposed project as impacts are predicted to be low beyond the boundary of the mine. In addition the peak concentrations are predicted to occur closer to the new activities. Recommendations Based on the findings of this assessment the following are recommended:

5 SRK Consulting: : East Dorstfontein AQIA Page iv Maintain the current monitoring network and where necessary make minor adjustments to accommodate the installation of new infrastructure from the proposed project. In instances where activities change the monitoring network may be adjusted accordingly. When the proposed project is operational an annual monitoring program should be maintained to determine whether the proposed project is having an impact on the surrounding environment with respect to dust fallout. A continuous PM 10 and PM 2.5 monitor should be installed at the mine or if possible at sensitive receptors in close proximity to the mine. Conduct periodic independent audits of monitoring systems and the implementation of management plans to ensure that the system is maintained and that suitable data is obtained for decision making. When fugitive dust can be observed leaving the area additional dust suppression should be applied to the affected areas. The following mitigation measures are proposed and may be considered, if practicable, for the new activities: Regular irrigation by water, especially during windy conditions at the site, of the access road and construction material and debris with just enough moisture to keep the dust down without creating runoff. Should water not be available as a result of drought conditions then chemical suppressants need to be considered. Reduction of speed on unpaved roads to reduce the entrainment of dust into the atmosphere. During grading activities, any exposed earth should be watered if it is going to be exposed for long periods, especially during windy conditions. On windy days, or when fugitive dust is dispersed from the Site of Works, additional application of water to the affected areas should be implemented. When and where applicable, soil stockpiles that will not be used should be re-vegetated as soon as possible, or kept wet during windy periods. During the operational phases for the proposed project any bare ground surrounding the main operational area but within the boundaries of the facility must be covered with suitable vegetation that will be able to grow in the area. When fugitive dust can be observed dispersing from the site, additional dust suppression should be applied to the affected areas. If possible, additional dust monitoring equipment needs to be installed in order to effectively monitor dust related impacts from the project area and thereafter dust emissions can be managed better. In places of high vehicular traffic on unpaved roads, dust suppression measures on the roads may be implemented to reduce dust levels from the entrainment of dust. These measures will range from watering of roads, application of a chemical dust suppressant where watering is impractical, and/or paving of roads. Reduce the possibility of spillage from vehicles by ensuring all loads are covered, for example, with tarpaulin.

6 SRK Consulting: : East Dorstfontein AQIA Page v Table of Contents Executive Summary... ii Disclaimer... ix List of Abbreviations... x 1 Introduction Introduction Background of the project Project Description Scope of report Project Identification Requirements Enterprise Details Project Location Emissions Characterisation Emission characteristics Existing emission sources Emission sources from proposed development Modelling scenarios Construction activities Operational activities Proposed emissions and source parameters Meteorological data Description of climate in southern Africa Climate conditions at the project site Rainfall Temperature Wind Field On-site weather station Lakes Environmental Data Summary of meteorological data Ambient impact analysis and ambient levels Ambient air quality legislation Air Quality Standards PM Dust fallout Ambient air quality monitoring data Dispersion modelling methodology Model Used Quantification of emissions... 44

7 SRK Consulting: : East Dorstfontein AQIA Page vi 6.3 Specify model settings Land use classification Surface Roughness Albedo Bowen Ration Description of modelled meteorological data Discussion on upper air data utilised Description of receptor grid Description of topographical data Dispersion modelling results PM PM Dust fallout Impact assessment Methodology Construction Operational phase Conclusions and Recommendations Conclusions Recommendations References Appendices Appendix A: Quantification of emissions... 68

8 SRK Consulting: : East Dorstfontein AQIA Page vii List of Tables Table 2-1: Enterprise details Table 2-2: Details of responsible person Table 2-3: Location and extent of East Dorstfontein Table 2-4: Nearest towns and villages to the project area Table 3-1: Summary of sources at East Dorstfontein Table 3-2: Emission rates for proposed operations Table 4-1: Average monthly rainfall from the on-site weather station and Lakes Environmental Table 4-2: Comparison between on-site and Lakes Environmental temperature data Table 5-1: Ambient air quality standards for PM Table 5-2: Ambient air quality standards for PM Table: 5-3: NEM: AQA, 2004 Acceptable Dust Fallout Rates given in mg/m 2 /day Table 5-4: Dust monitoring locations Table 5-5: Average monthly dust fallout concentrations for each year for the period March 2012 to February Table 6-1: Land types, -use, structures and vegetation cover Table 6-2: Surface roughness Table 6-3: Albedo Table 6-4: Bowen Ration Table 6-5: Receptors in close proximity to the site Table 7-1: Predicted PM 10 concentrations for the proposed project Table 7-2: Predicted PM 2.5 concentrations for the proposed project Table 7-3: Predicted dust fallout concentrations Table 8-1: Criteria for Assessing Significance of Impacts Table 8-2: Interpretation of Impact Rating Table 8-3: Impacts during the construction phase Table 8-4: Impacts during the operational phase... 63

9 SRK Consulting: : East Dorstfontein AQIA Page viii List of Figures Figure 1-1: Layout of East Dorstfontein Figure 2-1: Area map - Adjacent area (10 km radius from source) Figure 2-2: Regional map of the area (50 km radius from source) Figure 4-1: Annual variations in the positions of the South Atlantic and South Indian high pressure cells (Source: Preston-Whyte et al., 2000) Figure 4-2: Important features of the surface atmospheric circulation over southern Africa (Source: Preston- Whyte et al., 2000) Figure 4-3: Comparison of rainfall figures from the on-site weather station and Lakes Environmental Figure 4-4: Comparison of average Temperature data Figure 4-5: All hours, day time and night time wind roses (Source: Lakes Environmental) Figure 4-6: Seasonal wind roses (Source: Lakes Environmental) Figure 4-7: Wind class frequency distribution (Source: Lakes Environmental) Figure 4-8: All hours, day time and night time wind roses (Source: Lakes Environmental) Figure 4-9: Seasonal wind roses (Source: Lakes Environmental) Figure 4-10: Wind class frequency distribution (Source: Lakes Environmental) Figure 5-1: Location of monitoring points Figure 5-2: Monthly dust fallout concentrations at East Dorstfontein for the period March 2012 to February Figure 6-1: Receptor Grid: Figure 6-2: Terrain in the project area Figure 7-1: Maximum predicted 99 th percentile PM10 concentrations for the proposed project Figure 7-2: Maximum predicted Annual PM10 concentrations for the proposed project Figure 7-3: Maximum predicted 99 th percentile PM2.5 concentrations for the proposed project Figure 7-4: Maximum predicted annual PM2.5 concentrations for the proposed project Figure 7-5: Maximum predicted 99 th percentile PM2.5 concentrations for the proposed project... 59

10 SRK Consulting: : East Dorstfontein AQIA Page ix Disclaimer The opinions expressed in this Report have been based on the information supplied to SRK Consulting (South Africa) (Pty) Ltd (SRK) by Exxaro Coal Central (Pty) Ltd (Exxaro). The opinions in this Report are provided in response to a specific request from Exxaro to do so. SRK has exercised all due care in reviewing the supplied information. Whilst SRK has compared key supplied data with expected values, the accuracy of the results and conclusions from the review are entirely reliant on the accuracy and completeness of the supplied data. SRK does not accept responsibility for any errors or omissions in the supplied information and does not accept any consequential liability arising from commercial decisions or actions resulting from them. Opinions presented in this report apply to the site conditions and features as they existed at the time of SRK s investigations, and those reasonably foreseeable. These opinions do not necessarily apply to conditions and features that may arise after the date of this Report, about which SRK had no prior knowledge nor had the opportunity to evaluate.

11 SRK Consulting: : East Dorstfontein AQIA Page x List of Abbreviations AIR - Atmospheric Impact Report AQIA - Air quality Impact Assessment CO - Carbon Monoxide CO 2 - Carbon Dioxide DEA - Department of Environmental Affairs EA - Environmental Authorisation EAP - Environmental Assessment Practitioner EIA - Environmental Impact Assessment EMPR- Environmental Management Programme GN - Government Notice I&AP s - Interested and Affected Parties MPRDA - Minerals and Petroleum Resources Development Act Mamsl - meters above mean sea level NAAQS - National Ambient Air Quality Standards NEM AQA - National Environmental Management Air Quality Act NEMA - National Environmental Management Act NEM: AQA - National Environmental Management Air Quality Act NO 2 - Nitrogen Dioxide PM - Particulate Matter ROM - Run of mine SO 2 - Sulfur Dioxide US-EPA - United States Environmental Protection Agency VOC s - Volatile Organic Compounds

12 SRK Consulting: : Dorstfontein East AQIA Page 11 1 Introduction 1.1 Introduction SRK Consulting (South Africa) (Pty) Ltd (SRK SA) was appointed by Exxaro Coal Central (Pty) Ltd (Exxaro) to undertake the necessary Environmental Impact Assessment (EIA) in support of an application for an Environmental Authorisation (EA) for the expansion of its Dorstfontein East Mine in Mpumalanga. This report provides a baseline description of the air quality and meteorological conditions for the study area and assesses potential impacts associated with the proposed expansion may have on the air quality in the area of the mine. The outcomes of the impact assessment were used to identify management measures required to manage potential negative impacts. 1.2 Background of the Project Exxaro plans to undertake the necessary EIA process for the extension of Pit 1, the pipeline from Dorstfontein West to Dorstfontein East. This supplementary footprint does not fall within the existing Environmental Management Plan (EMP), thus, authorisation will be sought for the proposed infrastructure in terms of the National Environmental Management Act (Act No 107 of 1998) (NEMA) and National Water Act (Act No. 38 of 1996) (NWA). 1.3 Project Description Dorstfontein East is an opencast mine located approximately 18 km from Ga-Nala, formerly known as Kriel. The mine is located within the Emalahleni Municipality and began operations in Currently, only opencast mining is practised and underground mining is an option that is currently being investigated. Two opencast pits are currently mined on the eastern and western sides of the resource area with mining taking place both in the 2 and 4 coal seams. The current activities have been approved by both the Department of Mineral Resources (DMR) and the Department of Water and Sanitation (DWS). Dorstfontein Coal Mines (Pty) Ltd is now planning to extend its operations on the western side of the mine referred to as Pit 1 Extension. The area is part of the current approved mining right area but it has not been included in the current Environmental Management Programme Report and Life of Mine (LOM). 1.4 Scope of report In terms of Section 53(f) of NEM: AQA, the Department of Environmental Affairs (DEA), developed and published "Regulations Regarding Air Dispersion Modelling, 2014". The regulations were published in Government Gazette No on 11 July 2014 under Government Notice (GN) No The Code of Practice, contained in Appendix A of the regulations (of GN533 of Government Gazette No ), is prescribed as the technical Code of Practice for air dispersion modelling, and provides technical standards on the application of air dispersion models. The Code of Practice is applicable to: a) The development of an air quality management plan, as contemplated in Chapter 3 of the Act. b) The development of a priority area air quality management plan, as contemplated in Section 19 of the Act. c) The development of an Atmospheric Impact Report (AIR report), as contemplated in Section 30 of the Act.

13 SRK Consulting: : Dorstfontein East AQIA Page 12 d) The development of a specialist AQIA study, as contemplated in Section 37(2)(b) of the Act. In accordance with the application of the requirements of the Code of Practice to an AIR report or a specialist AQIA study (as per points c) and d) above), this assessment is submitted in accordance with the prescribed format Air Dispersion Modelling Study. The Scope of Report for this assessment will therefore follow the regulations prescribing the format of an Air Dispersion Modelling Study Report, and will provide all the information available and relevant in order to comply with the requirements of the regulations regarding air dispersion modelling (Code of Practice).

14 SRK Consulting: : Dorstfontein East AQIA Page 13 Figure 1-1: Layout of East Dorstfontein

15 SRK Consulting: : Dorstfontein East AQIA Page 14 2 Project Identification Requirements 2.1 Enterprise Details The details for the East Dorstfontein Mine are presented in Table 2-1 and the details of the relevant persons that can be contacted at the mine are in Table 2-2. Table 2-1: Enterprise details Criteria Enterprise Name Trading As Type of Enterprise, e.g. Company/Close Corporation/Trust Company/Close Corporation/Trust Registration Number (Registration Numbers if Joint Venture) Registered Address Postal Address Details Dorstfontein Coal Mines (Pty) Ltd Dorstfontein Coal Mines Private Company 1952/003176/07 Telephone Number (General) Fax Number (General) Industry Type/Nature of Trade Land Use Zoning as per Town Planning Scheme Land Use Rights if outside Town Planning Scheme Table 2-2: Details of responsible person Criteria Responsible Person Emission Control Officer Details Daniel Jacobus Chrisstofel Stapelberg William Seabi Telephone Number Cell Phone Number Fax Number Address N/A After Hours Contact Details Project Location William.Seabi@exxaro.com Private Bag X 5007 Ganala 2271 Private Bag X 5007 Ganala 2271 N/A Mining No Mining The location and extent of Dorstfontein East Coal Mine is presented in Table 2-3. The closest towns to the project are presented in Table 2-4. The activities taking place within 10 km of the mine are presented in Figure 2-1 and activities on a regional scale in Figure 2-2.

16 SRK Consulting: : Dorstfontein East AQIA Page 15 Table 2-3: Location and extent of Dorstfontein East Criteria Physical Address of the Mine Description of Site (Where No Street Address) Coordinates of Approximate Centre of Operations Extent (km²) Elevation Above Mean Sea Level (m) Province Metropolitan/District Municipality Local Municipality Designated Priority Area (if applicable) Details N/A Plant and associated structure, opencast operations (Pit 1, 2, 3), discard facility, roads and conveyors, water pipelines, clean and dirty water infrastructure, power supply, sewage treatment works and a railway line and associated power supply 26 11' 37" S 29 21' 13" E 17.2 km² 1581 m Mpumalanga Nkangala District Municipality Emalahleni Local Municipality Highveld Priority Area The current land uses in the region include coal mining, farming and power generation. The major town in close proximity to the mine include Witbank with smaller towns such as Middelburg, Bethal and Kriel and Ogies. In addition, there are various farms and homesteads surrounding the mining area. Besides the farmers, there are numerous farm workers residing on the farms. Table 2-4: Nearest towns and villages to the project area Town/Village Distance (km) Direction Kriel 10.4 Southwest Hendrina 37.1 East Ogies 34.1 Northwest Springbok 16.8 North Bethal 30.6 Southwest Emalahleni 37.5 North Reedstream Park 17.2 West-northwest Thubelihle 6.0 West-southwest Evander 38.2 South-southwest Secunda 38.1 Southwest

17 SRK Consulting: : Dorstfontein East AQIA Page 16 Figure 2-1: Area map - Adjacent area (10 km radius from source)

18 SRK Consulting: : Dorstfontein East AQIA Page 17 Figure 2-2: Regional map of the area (50 km radius from source)

19 SRK Consulting: : Dorstfontein East AQIA Page 18 3 Emissions Characterisation 3.1 Emission Characteristics Identifying the sources that contribute to airborne particulate emissions in the project area is important for establishing sources of air pollution and cumulative impacts associated within the East Dorstfontein Mining Area, and the impact that the activities at the facility could have on air quality Existing Emission Sources The project area and surrounding land can be described as rural/industrial with large scale industrial activities in the area. The area is characterised by one large town (Witbank), smaller towns such as Ogies, Bethal, Kriel, and smaller settlements and farms in the area. The following sources of air emissions have been identified in the area: Mining activities. Power generation. Vehicle emissions. Fugitive dust sources (windblown dust especially during the dry season). Farming activities such as land preparation and harvesting. Biomass burning Mining Activities Numerous coal mines and activities at these mines such as drilling and blasting, crushing, hauling, and materials handling all contribute to pollutants in the atmosphere. The main pollutant of concern is particulate matter as this dust is inhalable and can cause respiratory illnesses. The main mines that are within 50km of East Dorstfontein are: New Clydesdale Coal. Douglas Colliery. Goedehoop Colliery. Ilanga Colliery. Dorstfontein Coal Mine. Arthur Taylor Colliery. Polmaise Colliery. Anglo Coal SA (Pty) Ltd Kriel. Waterpan Colliery South Witbank Colliery. Koorfontein Mines. Siyanda. Coal Kleinkopje Colliery. Exxaro Coal (Pty) Ltd Matla Colliery. Power Generation Coal is the main product mined in the area and coal power stations have been built in close proximity to these mines. These power stations supply electricity to the national power grid. The main pollutants emitted from these power stations are sulfur dioxide (SO 2), carbon monoxide (CO), carbon dioxide (CO 2), nitrogen dioxide (NO 2) and particulate matter (PM). The closest power stations to East Dorstfontein are: Matla (23 km southwest).

20 SRK Consulting: : Dorstfontein East AQIA Page 19 Kriel (18 km southwest). Duvha (27km north). Komati (17 km northeast). Kendal (40 km northwest).and Hendrina (31 km northeast). Vehicle Emissions Vehicle tailpipe emissions are always present and depending on whether the vehicle is maintained efficiently, the tailpipe emissions can contribute heavily or minimally to air pollution. Vehicle emissions can be classified into two groups, namely, primary and secondary pollutants. Pollutants such as carbon monoxide (CO), carbon dioxide (CO 2), sulfur dioxide (SO 2), oxides of nitrogen (NO x), particulates and lead are generally released into the atmosphere depending on the type of fuel that is used. These pollutants are termed primary pollutants. Secondary pollutants exist only because of the chemical reactions that take place in the atmosphere. Pollutants formed during this process include nitrogen dioxide (NO 2), photochemical oxidants (e.g. ozone), sulfates and nitrates. Vehicle tailpipe emissions are expected to be medium-high due to the relatively vehicle usage in the area. Fugitive Dust Sources All pollutants that arise from fugitive dust sources are termed primary pollutants as they are unlikely to undergo any physical or chemical reactions. Fugitive dust sources can be the vehicle entrainment of dust from gravel or unpaved roads and wind erosion of open areas. Particulate emissions from roadways depend on the road and the number of vehicles using the road. Windblown dust resulting from the erosion of bare ground depends on the velocity of the wind, the size of the exposed area and moisture and silt content of such areas. Areas that receive high amounts of rainfall will experience lower levels of fugitive dust being released into the air as higher moisture content makes the soil heavier and more compact, resulting in the soil being more resistant to wind erosion. At Dorstfontein East, fugitive dust emissions are expected to be low as the rainy season is from October to March and the wind speeds are generally low during the winter months. However, based on dust fallout monitoring results, there are instances where higher dust fallout concentrations are observed in the wet months when compared to the drier months. Fugitive dust from the roads network in the area is expected to be low-medium as a result of the frequency and number of vehicles that use the roads in the area. Farming Activities such as Land Preparation and Harvesting Farming is one of the main activities that occurs in the project area and include -land preparation and harvesting. Tractors are used to prepare large areas of land for cultivation. Dust emissions related to this activity is expected to be high as soils are loosened and suspended thereby allowing wind to easily transport the finer particles. The remaining vegetation is burned prior to the preparation of land and gases such as SO 2, NO x, CO and CO 2 are released from this activity. Dust is also generated during the harvesting activity and this adds to the ambient dust load. The fields are left bare at the end of the harvesting and before the preparation of land and this could increase windblown dust in the area, especially during the warmer months when wind speeds tend to increase. However lost dust emissions are expected during the wet season Biomass Burning The burning of crop residue and veld fires are sources of emissions that can be associated with areas that are densely vegetated. Biomass burning, such as the grasslands in Witbank, is significant as these areas are cleared for agricultural purposes by slash and burn methods. Biomass burning is an incomplete combustion process with carbon monoxide, methane and nitrogen dioxide emitted during the process. Biomass burning in the local area and regionally could negatively impact on air quality, albeit on a seasonal basis.

21 SRK Consulting: : Dorstfontein East AQIA Page Emission Sources from Proposed Development This assessment is based on plans for the construction and operational phases of the proposed development, notably the proposed extension of the opencast mining area. This assessment represents a conservative (worst case, full production) approach to the evaluation of the impact of potential emissions arising from the activities. Construction activities include the following: Clearing of land i.e. topsoil and vegetation. Grading and bulldozing. Haulage of topsoil and vegetation. Vehicle entrainment of dust on access roads. Stockpiling of topsoil. Materials handling of topsoil. Vehicle tailpipe emissions. Operations activities for the project include the following: Mining at the open pit. Mining at the open pit may be undertaken by the following methods (that are still to be investigated): In Pit Crushing with underground feeder breakers and Conveying; Dragline Overburden Removal; and Truck and Shovel Operations (preferred alternative). The various mining processes in the general operational cycle will include: Top soil removal; Drill and blast of the hard overburden; Dozing; Loading (truck and shovel or dragline); Coal removal (blast and haul); Transporting (hauling or conveying); Placing (dumping and/or spreading); and Rehabilitation. Road transport of coal from the pit to the plant. Materials handling. Coal storage at the ROM stockpile. On-site vehicle movements associated with mining activities. Vehicle tailpipe emissions. Based on observations during the site visit in February 2016, dust is considered to be the predominant pollutant type that will be generated from the various activities listed above. Materials handling, wind erosion (e.g. discard dumps, coal stockpiles, roads) and vehicle entrainment of dust from unpaved roadways are expected to be the main sources of dust in the area during the operational phase of the mine.

22 SRK Consulting: : Dorstfontein East AQIA Page 21 Table 3-1: Summary of Pollutant Sources at and in the vicinity of East Dorstfontein Mine Activity Timing Pollutant Type Clearing of land and excavation Haulage of material Windblown dust from stockpiles Mining activities Comment Construction Dust The clearing of land will be a source of dust as winds may blow dust off bare ground. Construction Dust Dust may be blown off haul trucks transporting excavated material to stockpile areas. Construction and operational Current, construction, operational and post operational Plant Current and operational Roads & motor vehicles Current, construction, operational, and post-operational Dust Dust Dust Dust, SO 2, NOx, Volatile Organic Compounds (VOCs) Whilst the land is being cleared, it is highly likely that soil will be stockpiled. Wind erosion of the stockpiles will occur resulting in the dispersion of dust. Mining activities will include drilling and blasting, tipping and crushing. Hauling of coal from the opencast mine to the plant are also part of the mining activities. Tipping, loading, crushing and screening are the main processes at the plant. Most of the process are wet, therefore, dust emissions are expected to be low. Source of particulate matter resulting from the resuspension of dust from the unpaved road surfaces. Since there is a relatively low vehicle count in the area, this is currently considered to be a minor source of air pollution. Dust will be reduced further when hauling of coal by trucks to the discard dump stops and a conveyor will be constructed instead. Diesel vehicle will also be a source of gaseous (SO 2 and NO x) and particulate emissions within the mine area. ROM stockpile Operational and operational Dust Low fine particle content and hence not a major dust source. Dust will be limited to active areas of the stockpiles. Materials handling Current, Operational Dust The handling of coal is the main dust generating activity at the mine. Fine particulate matter can be easily generated when handling the coal. The coal material has limited erodibility, however, as the movement of coal between the mine, plant and discard dump becomes more frequent, dust will easily be generated from this activity as the coal materials collide together. Farming Current, Operational, Post Operational Dust Dust is generated through the clearing of land and harvesting. Burning of vegetation will release gases such as SO 2, NO x, CO and CO 2. Bare ground after harvesting is a source of windblown dust in the area especially during the months of August when wind speeds tend to increase.

23 SRK Consulting: : Dorstfontein East AQIA Page Modelling Scenarios Construction Activities No dispersion model model was simulated for dust emissions during the preconstruction/construction phase as activities undertaken during these phases are short-lived in nature and the entire phase has short term duration. Hence, emissions during this phase are deemed to be low to negligible and short lived in nature Operational Activities The models were set up based on the project description for the proposed activities and only new infrastructure i.e. opencast pit, materials handling (Discard dump and open pit) and haul roads were modelled. Two scenarios were run, a scenario without the implementation of management measures and a scenario with the implementation of management measures for activities such as the vehicle entrainment of dust from the haul roads. The scenarios are outlined as follows: Scenario 1 without management measures: For the proposed operational conditions dust emissions were modelled for open pits, materials handling and the haul road. The dust sources that were modelled during the operational phase were without any management measures in place and, hence, this scenario is viewed as the worst case scenario. No chemical suppressant or chemical sprayers will be used on the haul road. Scenario 2 with management measures: It is assumed that, during the proposed operational conditions, management measures will be used to reduce dust emissions within the mine license boundary. Dust emissions from the haul road are suppressed due to rainfall (as natural mitigation) in the area as a form of natural mitigation as well as wet suppression through water bowsers. The emission rates for each source for each scenario are presented in Table Proposed Emissions and Source Parameters Table 3-2: Emission rates for proposed operations Source Scenario 1 without management Scenario 2 with management PM 10 PM 2.5 Dust fallout PM 10 PM 2.5 Dust fallout Units tpa tpa tpa tpa tpa tpa Open Pit Haul Road Materials handling (Truck Dumping) Materials handling (Bulldozer) Materials handling (FEL) Discard dump (active stockpile)

24 SRK Consulting: : Dorstfontein East AQIA Page 23 4 Meteorological Data 4.1 Description of Climate in Southern Africa South Africa has a temperate climate with warm sunny days occurring most of the year. South Africa s summer runs from November to February when most of the country is characterised by hot weather and afternoon thunderstorms. The winter season, which occurs from May to August, is usually mild and dry except for the south-western parts of the country. South Africa is influenced by two oceanic currents, the warm south flowing Mozambique-Agulhas current and the north flowing Benguela current. These two currents have an influence on the climate and vegetation between the eastern and western sides of the country. The east coast generally has the higher temperatures and rainfall when compared to the west coast. Climate conditions in South Africa range from Mediterranean in the southwestern part of the country to temperate in the interior plateau and subtropical in the north-east. There is also evidence of a desert climate in the north-western part of the country. South Africa is located in the subtropical high pressure belt, where subsidence, high pressure and atmospheric stability are predominant features. Three high pressure (HP) cells dominate over South Africa. The South Atlantic HP cell is located off the west coast, the Continental HP cell reigns over the interior of the country and the South Indian HP cell resides over the east coast. This results in the annual mean circulation of the atmosphere over South Africa being anticyclonic (Steynor, 2006). During winter, the anticyclonic circulation over South Africa is well established and is at its most intense. The Continental HP cell migrates north over the country and the South Indian HP cell shifts westward towards the east coast from the summer position of 88 E to 65 E in winter. The South Atlantic HP cell moves from its summer position at 4 W to its winter position at 16 W, migrating away from the western coast (Preston-Whyte et al., 1976). Anticyclones result in subsidence in the atmosphere, which increase the incidence and duration of calm winds, clear skies, lowered humidity and little precipitation (Preston-Whyte et al., 1980) (Figure 4-1). East Dorstfontein Air Quality Study Annual variations Project No Figure 4-1: Annual variations in the positions of the South Atlantic and South Indian high pressure cells (Source: Preston-Whyte et al., 2000)

25 SRK Consulting: : Dorstfontein East AQIA Page 24 Conversely, during summer, a low pressure system dominates over the interior of the country due to the slight southward shift of the continental HP cell (Figure 4-2). The South Atlantic HP cell moves towards the east, over the Western Cape, and the South Indian HP shifts eastwards, causing the high pressure conditions over the eastern coast to diminish. Extremely stable atmospheric conditions that can persist for long periods are the result of the semi-permanent and subtropical continental anticyclones. These are found to occur at a frequency of 70% and 20% in winter and summer, respectively. East Dorstfontein Air Quality Study Surface atmospheric ciculation Project No Figure 4-2: Important features of the surface atmospheric circulation over southern Africa (Source: Preston-Whyte et al., 2000) The south-easterly trade winds generally affect South Africa throughout the year. However, during winter, the high pressure cells shift northward, causing the circumpolar westerly waves to displace the easterly trade winds and dominate over South Africa. The westerly belt is associated with the migration of isolated low pressure cells and cyclones around the coast or across the country towards the east. During summer, the high pressure belt shifts southwards and the easterly trade winds displace the westerly waves to resume its influence over the country. During summer, localised weather systems to the east of the south-easterly trade winds causes turbulence and uplift and the potential for precipitation over the eastern part of the country, resulting in summer rains. On the western side of the easterly waves, upper-level convergence and surfacelevel divergence causes clear conditions with no precipitation over the western part of the country. During winter, westerly waves significantly influence the weather of the country. Upper-level divergence and surface-level convergence occurs to the rear of the trough, which causes uplift and cloud formation resulting in precipitation and winter rains over the western coast. Rainfall will also occur with the passing of cold fronts, which are associated with the westerly waves. Rainfall has a positive effect on pollution control as the water droplets act as nuclei onto which dust and pollutants will collect and deposit onto the ground. This is known as scrubbing of the atmosphere.

26 SRK Consulting: : Dorstfontein East AQIA Page 25 Along the coastline, sea and land breeze circulations influence the diurnal wind variation and ultimately govern the transport of atmospheric pollutants. During the daytime, the land heats rapidly while the sea retains its cool temperature. The warm air over the land rises causing a low pressure to develop. The cool air over the sea subsides and flows down the pressure gradient, causing a sea-land breeze to develop. The converse is true for night time conditions, where the air above the land cools due to a lack of insulation, while the air above the sea remains warm. A land-sea breeze will therefore prevail during the night (Diab, pers comm., 2007). 4.2 Climate Conditions at the Project Site According to the Koppen Climate Classification System the project area is classified as the Category C climate type which is characteristically a moist mid latitude climate with mild winters. Annual precipitation is less than 760 mm. The project site itself falls into subcategory Cwb within the C category. The Cwb climate type can be classified as mild temperate with dry winters and warm summers. Weather data for the project site has been acquired from the on-site weather station for the period January 2015 to February 2017 and from Lakes Environmental for the period January 2013 to December Rainfall Rainfall is an important parameter with respect to air quality. During the rainy season, air pollution, and more specifically in this case, dust particles, are removed from the atmosphere. Dust emissions are suppressed due to increases in soil moisture content and increased vegetation cover during the rainy season. During the cooler dry and hot dry seasons, dust emission levels are generally higher. The average annual rainfall measured at the on-site weather station for the period from January 2015 to February 2017 is 471 mm (Table 4-1). The majority of rainfall is received from October to March. There is low to negligible rainfall from April to September. The month with the highest average rainfall is November (124.5 mm). The average annual modelled rainfall from Lakes Environmental for the period from January 2013 to December 2015 is mm (Table 4-1). The majority of rainfall is received from October to March. There is low to negligible rainfall from April to September. The month with the highest average rainfall is December (145.8 mm). The average monthly rainfall is presented in Figure 4-3.

27 SRK Consulting: : Dorstfontein East AQIA Page 26 Table 4-1: Average monthly rainfall from the on-site weather station and Lakes Environmental Month On-site weather station (Jan15 to Feb17) Lakes Environmental (Jan13 to Dec15) Units mm mm January February March April May June July August September October November December Total

28 SRK Consulting: : Dorstfontein East AQIA Page Comparison of rainfall from the on-site weather station and Lakes Environmental Rainfall (mm) January February March April May June July August September October November December Month On-site rainfall Lakes Environmental On-site cumulative rainfall Lakes Environmental cumulative rainfall Figure 4-3: Comparison of rainfall figures from the on-site weather station and Lakes Environmental

29 SRK Consulting: : Dorstfontein East AQIA Page Temperature Similar to rainfall, temperature data has been acquired from the on-site weather station and modelled temperature data has been sourced from Lakes Environmental. The monthly average, highest maximum and lowest minimum temperatures are presented in the Table 4-2 and the daily average temperatures are presented in Figure 4-4. The data from the on-site weather station shows average temperatures range from C with maximum temperatures reaching a high of 44 C (February) and minimum temperatures at a low of C (June). The data from Lakes Environmental shows average temperatures range from C with maximum temperatures reaching a high of 30.1 C (December) and minimum temperatures at a low of -3.4 C (July). Table 4-2: Comparison between on-site and Lakes Environmental temperature data Month On-site weather station Lakes Environmental Average Maximum Minimum Average Maximum Minimum January February March April May June July August September October November December

30 SRK Consulting: : Dorstfontein East AQIA Page Comparison of temperature data 20.0 Temoerature ( C) January February March April May June July August September October November December Month Onsite weather station Lakes Environmental Figure 4-4: Comparison of average daily temperature data

31 SRK Consulting: : Dorstfontein East AQIA Page Wind Field The wind field for an area is an important parameter with respect to air quality and winds can generate dust emissions as well as control the dispersion of an emissions plume. The degree to which winds can influence dispersion depends on the wind speed. Higher wind speeds result in longer travel distance and dilution of the pollutants and lower, more stable wind conditions result in shorter travel distance and build-up of pollutant levels (especially gases) over a smaller area On-Site Weather Station Wind meteorological data from the on-site weather station for the period January 2015 to February 2017 was used to generate wind roses. The wind roses for all hours, day and night time for the site for the period January 2015 to February 2017 are presented in Figure 4-5. The prevailing winds are relatively constant throughout the year in the project area. The prevailing winds for this period are from the north, west-southwest and southeast, with lower occurrences from the east, east-southeast and south-southeast (Figure 4-5a). Wind patterns observed during the day (Figure 4-5b) and earlier parts of the night (18h00-00h00) (Figure 4-5) are similar, however, with winds from the southeast becoming more pronounced during the latter period. Winds during the latter parts of the night (00h00-06h00) the prevailing winds are from the westsouthwest with lower occurrences from the southwest, southeast, north and west. The average wind speed measured for all hours is 4.47 m/s with maximum speeds less than 11.1 m/s and with calms (<0.5 m/s) of 1.72%. The average wind speeds for daytime (06h00-18h00) during the year is 5.28 m/s with calms of 1.22%. The average wind speed during the earlier parts of the night (18h00-23h00) is 4.14 m/s with calms of 0.45% and the average wind speed decreases during the latter parts of the night to 3.05 m/s with calms prevailing 4.21% of the time. Seasonal wind roses were also created for the period (Figure 4-6). The main prevailing wind directions during the seasons general resemble the annual wind rose with the most variation in wind directions and higher wind speeds occurring during spring. The highest average wind speeds of 5.56 m/s occurs during the spring with calms prevailing 0.53% of the time. The lowest average wind speeds occur during winter with an average wind speed of 3.77 m/s with calm conditions prevailing 4.19% of the time. The winds during summer and autumn are 4.65 and 3.87 m/s respectively. Figure 5-5 represents the wind class frequency distribution which shows that 70.8% of wind speeds are below 5.7 m/s (low speeds) with 29.2% of winds above 5.7 m/s (high speeds).

32 WIND ROSE PLOT: COMMENTS: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: SRK Consulting: : Dorstfontein East AQIA Page 31 East Dorstfontein Coal Mine January 2015 to February All hours Wind Speed Direction (blowing from) East Dorstfontein Coal Mine January 2015 to February Day Time 06h00-18h00 Wind Speed Direction (blowing from) NORTH NORTH 25% 30% 20% 24% 15% 18% 10% 12% 5% 6% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) >= >= SOUTH SOUTH Calms: 1.72% Calms: 1.22% Start Date: 2015/01/01-00:00 End Date: 2017/02/23-23:00 SRK Consulting (South Africa) Pty Ltd Start Date: 2015/01/01-06:00 End Date: 2017/02/23-18:00 SRK Consulting (South Africa) Pty Ltd Dhiren Naidoo Dhiren Naidoo 1.72% hrs. 1.22% hrs m/s 2017/03/ m/s 2017/03/ Figure 4-5a: January 2013 to December 2015 (All hours) Figure 4-5b: January 2013 to December 2015 (Day Time 06h00-18h00) East Dorstfontein Coal Mine January 2015 to February Night Time 18h00-23h00 Wind Speed Direction (blowing from) East Dorstfontein Coal Mine January 2015 to February Night Time 00h00-06h00 Wind Speed Direction (blowing from) NORTH NORTH 20% 20% 16% 16% 12% 12% 8% 8% 4% 4% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) >= >= SOUTH SOUTH Calms: 0.45% Calms: 4.21% Start Date: 2015/01/01-18:00 End Date: 2017/02/23-23:00 SRK Consulting (South Africa) Pty Ltd Start Date: 2015/01/01-00:00 End Date: 2017/02/23-06:00 SRK Consulting (South Africa) Pty Ltd Dhiren Naidoo Dhiren Naidoo 0.45% 4710 hrs. 4.21% 5459 hrs m/s 2017/03/ m/s 2017/03/ Figure 4-5c: January 2013 to December 2015 (Night Time 18h h00) Figure 4-5d: January 2013 to December 2015 (Night Time 00h00-06h00) East Dorstfontein Air Quality Study On-site all hours, day time and night time wind roses Project No Figure 4-5: All hours, day time and night time wind roses (Source: Lakes Environmental)

33 WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: SRK Consulting: : Dorstfontein East AQIA Page 32 East Dorstfontein Coal Mine Spring (All hours) Wind Speed Direction (blowing from) East Dorstfontein Coal Mine Summer (All hours) Wind Speed Direction (blowing from) NORTH NORTH 20% 25% 16% 20% 12% 15% 8% 10% 4% 5% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) >= >= SOUTH SOUTH Calms: 0.53% Calms: 0.48% Start Date: 2015/09/01-00:00 End Date: 2016/11/30-23:00 SRK Consulting (South Africa) Pty Ltd Start Date: 2015/01/01-00:00 End Date: 2017/02/23-23:00 SRK Consulting (South Africa) Pty Ltd Dhiren Naidoo Dhiren Naidoo 0.53% 4355 hrs. 0.48% 5624 hrs m/s 2017/03/ m/s 2017/03/ Figure 4-6a: Spring (All hours) Figure 4-6b: Summer (All hours) East Dorstfontein Coal Mine Autumn (All hours) Wind Speed Direction (blowing from) East Dorstfontein Coal Mine Winter (All hours) Wind Speed Direction (blowing from) NORTH NORTH 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) >= >= SOUTH SOUTH Calms: 2.00% Calms: 4.19% Start Date: 2015/03/01-00:00 End Date: 2016/05/31-23:00 SRK Consulting (South Africa) Pty Ltd Start Date: 2015/06/01-00:00 End Date: 2016/08/31-23:00 SRK Consulting (South Africa) Pty Ltd Dhiren Naidoo Dhiren Naidoo 2.00% 4411 hrs. 4.19% 4414 hrs m/s 2017/03/ m/s 2017/03/ Figure 4-6c: Autumn (All hours) Figure 4-6d: Winter (All hours) East Dorstfontein Air Quality Study Seasonal wind roses Project No Figure 4-6: Seasonal wind roses (Source: Lakes Environmental)

34 SRK Consulting: : Dorstfontein East AQIA Page 33 East Dorstfontein Air Quality Study Wind class frequency distribution Project No Figure 4-7: Wind class frequency distribution (Source: Lakes Environmental) Lakes Environmental Data Modelled hourly meteorological data for the period January 2013 to December 2015 was acquired from Lakes Environmental. The wind roses for all hours, day and night time for the site for the period January 2013 to December 2015 are presented in Table 4-8. The prevailing winds are relatively constant throughout the year in the project area. The prevailing winds for this period are from the north-northwest, northwest and north, with lower occurrences from the east, east-northeast and east-southeast (Figure 4-8a). Wind patterns observed during the day (Figure 4-8) and earlier parts of the night (18h00-23h00) (Figure 4-8) are similar, however, with winds from the east becoming more pronounced at night. Prevailing winds during the latter parts of the night (00h00-06h00) are become predominantly easterly, eastnortheasterly and northerly with lower occurrences from the northeast and east-southeast. The average wind speed measured for all hours is 2.93 m/s with maximum speeds less than 11.1 m/s and with calms (<0.5 m/s) of 10.98%. The average wind speeds for daytime (06h00-18h00) during the year is 2.83 m/s with calms of 12.84%. The average wind speed during the earlier parts of the night (18h00-23h00) is 3.05 m/s with calms of 8.52% and the average wind speed decreases during the latter parts of the night to 2.93m/s with calms prevailing 8.83% of the time. Seasonal wind roses were also created for the period (Figure 4-9). The main prevailing wind directions during spring resemble the annual wind rose. However, during summer the prevailing winds are from the north, north-northwest, north-northwest and east with a lower occurrence from the northeast, northnortheast and east-southeast. The prevailing winds for autumn and winter are similar, however there are a higher frequency of winds from the west during autumn when compared to winter. However it should be noted that winds speeds during winter are higher when compared to autumn. The highest average wind speeds of 3.54 m/s occurs during the spring with calms prevailing 5.63% of the time.

35 SRK Consulting: : Dorstfontein East AQIA Page 34 The lowest average wind speeds occur during autumn with an average wind speed of 2.46m/s with calm conditions prevailing 15.81% of the time. The winds during summer and winter are 2.95 and 2.76 m/s respectively. Figure 5-5 represents the wind class frequency distribution which shows that 93.2% of wind speeds are below 5.7 m/s (low speeds) with 6.8% of winds above 5.7 m/s (high speeds).

36 WIND ROSE PLOT: COMMENTS: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: SRK Consulting: : Dorstfontein East AQIA Page 35 East Dorstfontein Coal Mine 1 January 2013 to 31 December 2015 Wind Speed Direction (blowing from) East Dorstfontein Coal Mine 1 January 2013 to 31 December 2015 (Day Time 06h00-18h00) Wind Speed Direction (blowing from) NORTH NORTH 10% 15% 8% 12% 6% 9% 4% 6% 2% 3% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) SOUTH >= SOUTH >= Calms: 10.98% Calms: 12.84% Start Date: 1/1/ :00 End Date: 12/31/ :00 SRK Consulting Start Date: 1/1/ :00 End Date: 12/31/ :00 SRK Consulting D. Naidoo D. Naidoo 10.98% hrs % hrs m/s 8/19/ m/s 8/19/ Figure 4-8a: January 2013 to December 2015 (All hours) Figure 4-8b: January 2013 to December 2015 (Day Time 06h00-18h00) East Dorstfontein Coal Mine 1 January 2013 to 31 December 2015 (Night Time 18h00-23h00) Wind Speed Direction (blowing from) East Dorstfontein Coal Mine 1 January 2013 to 31 December 2015 (Night Time 00h00-06h00) Wind Speed Direction (blowing from) NORTH NORTH 15% 15% 12% 12% 9% 9% 6% 6% 3% 3% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) SOUTH >= SOUTH >= Calms: 8.52% Calms: 8.83% Start Date: 1/1/ :00 End Date: 12/31/ :00 SRK Consulting Start Date: 1/1/ :00 End Date: 12/31/ :00 SRK Consulting D. Naidoo D. Naidoo 8.52% 6570 hrs. 8.83% 7665 hrs m/s 8/19/ m/s 8/19/ Figure 4-8c: January 2013 to December 2015 (Night Time 18h00-23h00) Figure 4-8d: January 2013 to December 2015 (Night Time 00h00-06h00) East Dorstfontein Air Quality Study On-site all hours, day time and night time wind roses Project No Figure 4-8: All hours, day time and night time wind roses (Source: Lakes Environmental)

37 WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software WIND ROSE PLOT: COMMENTS: WRPLOT View - Lakes Environmental Software DATA PERIOD: CALM WINDS: AVG. WIND SPEED: DATA PERIOD: CALM WINDS: AVG. WIND SPEED: COMPANY NAME: MODELER: TOTAL COUNT: DATE: COMPANY NAME: MODELER: TOTAL COUNT: DATE: DISPLAY: DISPLAY: PROJECT NO.: PROJECT NO.: SRK Consulting: : Dorstfontein East AQIA Page 36 East Dorstfontein Coal Mine Spring (All hours) Wind Speed Direction (blowing from) East Dorstfontein Coal Mine Summer (All hours) Wind Speed Direction (blowing from) NORTH NORTH 15% 15% 12% 12% 9% 9% 6% 6% 3% 3% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) SOUTH >= SOUTH >= Calms: 5.63% Calms: 7.52% Start Date: 9/1/ :00 End Date: 11/30/ :00 SRK Consulting Start Date: 1/1/ :00 End Date: 12/31/ :00 SRK Consulting D. Naidoo D. Naidoo 5.63% 6552 hrs. 7.52% 6480 hrs m/s 8/19/ m/s 8/19/ Figure 4-9a: Spring (All hours) Figure 4-9b: Summer (All hours) East Dorstfontein Coal Mine Autumn (All hours) Wind Speed Direction (blowing from) East Dorstfontein Coal Mine Winter (All hours) Wind Speed Direction (blowing from) NORTH NORTH 10% 15% 8% 12% 6% 9% 4% 6% 2% 3% WEST EAST WEST EAST WIND SPEED (m/s) WIND SPEED (m/s) SOUTH >= SOUTH >= Calms: 15.81% Calms: 14.82% Start Date: 3/1/ :00 End Date: 5/31/ :00 SRK Consulting Start Date: 6/1/ :00 End Date: 8/31/ :00 SRK Consulting D. Naidoo D. Naidoo 15.81% 6624 hrs % 6624 hrs m/s 8/19/ m/s 8/19/ Figure 4-9c: Autumn (All hours) Figure 4-9d: Winter (All hours) East Dorstfontein Air Quality Study Seasonal wind roses Project No Figure 4-9: Seasonal wind roses (Source: Lakes Environmental)

38 SRK Consulting: : Dorstfontein East AQIA Page 37 East Dorstfontein Air Quality Study Wind class frequency distribution Project No Figure 4-10: Wind class frequency distribution (Source: Lakes Environmental) 4.4 Summary of Meteorological Data Based on the meteorological data from the on-site weather station the climate in the project area is expected to be seasonal, with distinct warm months (September to February) and cool months (March to August). Rainfall is higher during the months from October to April and lower from May to September. The prevailing winds for this period are from the north, west-southwest and southeast, with lower occurrences from the east, east-southeast and south-southeast as observed on the site windroses. The average wind speed measured for all hours is 4.47 m/s with maximum speeds less than 11.1 m/s and with calms (<0.5 m/s) of 1.72%. 5 Ambient Impact Analysis and Ambient Levels 5.1 Ambient Air Quality Legislation In South Africa, the main legislation governing air quality is the National Environment Management: Air Quality Act, No. 39 of 2004 (NEM: AQA). An important factor in the approach to air quality has been the promulgation of regulations with respect to ambient air quality standards under NEM: AQA. These standards provide the goals for air quality management plans and the context within which the effectiveness of these management plans are measured. The NEM: AQA identifies priority pollutants, determination of priority areas affected by these pollutants and the setting of ambient standards with respect to these pollutants within these areas. This Section provides standards for priority pollutants i.e. PM 10, SO 2 etc. All ambient air quality monitoring data will be compared against the South African National Ambient Air Quality Standards (NAAQS) that came into effect on 24 December 2009 (Notice No.: Government Gazette No December 2009).

39 SRK Consulting: : Dorstfontein East AQIA Page Air Quality Standards PM10 Particulate matter (PM) is airborne particles that include dust, smoke and soot. Particulate matter can either be emitted naturally (e.g. windblown dust of stockpiles) or through human activity (e.g. stack emissions). It is defined by size, with coarse particles being between microns, fine particles less than 2.5 microns, and ultrafine particles less than 0.1 microns in diameter. Particulate matter can have adverse effects on humans such as respiratory illnesses (asthma and bronchitis) or cardiovascular diseases. It can also affect vegetation in two ways, namely, by inhibiting the plant s photosynthetic properties by coating the leaves thereby blocking the penetration of natural light. Furthermore deposition onto soils of various metals that could be in the particulate matter can be which are absorbed by vegetation thereby hindering plant growth. The uptake of metals by plants also has the potential to contaminate vegetables and fruit that may be consumed by humans and animals. Ambient air quality standards for PM 10 and PM 2.5 are presented in Table 5-1 and Table 5-2 respectively. PM 10 and PM 2.5 are important as it provides a measure of respirable dust, which has the potential to affect human health. Table 5-1: Ambient air quality standards for PM 10 Standard 24-hour Annual Average Units µg/m 3 µg/m 3 South African Standard (Effective from 1 January 2015) Frequency of exceedance 4 0 1) As listed in the NEM: AQA. Government Gazette No December 2009 Table 5-2: Ambient air quality standards for PM 2.5 Standard 24- hour Annual Average Units µg/m 3 µg/m 3 South African Standard (Effective from 1 January 2016 to 31 December 2029) Frequency of exceedance 4 0 1) As listed in the NEM: AQA. Government Gazette No June Dust fallout This Section provides standards for priority pollutants i.e. dust fallout. The National Dust Control Regulations came into effect on 1 November In terms of the NEM: AQA the acceptable dust fallout rates are presented in Table 5-3. Table: 5-3: NEM: AQA, 2004 Acceptable Dust Fallout Rates given in mg/m 2 /day Restriction Area Dust fall rate (D) (mg/m 2 /day, 30-day average) Permitted frequency of exceeding dust fall rate Residential Area D < 600 Two within a year, not sequential months Non-residential Area 600 < D < 1,200 Two within a year, not sequential months Two exceedances of the Residential Area and Non-residential area are permitted within a year, but not for two sequential months. If there are more than two exceedances within a year, or two in

40 SRK Consulting: : Dorstfontein East AQIA Page 39 consecutive months, then within three months after the submission of the dust fallout monitoring report a dust management plan must be submitted to the air quality office for approval. 5.3 Ambient Air Quality Monitoring Data Dust fallout monitoring is being undertaken at the East Dorstfontein Mine. SRK received monitoring data for the period March 2012 to February 2017 from the mine. Dust buckets have been added to the network over the course of the monitoring period. The monitoring locations are presented in Table 5-4 and Figure 5-1. The dust fallout monitors installed in and around the mining area are classified as Residential Area and Non-residential Area dust buckets in accordance with the National Dust Control Regulations. All dust bucket locations are classified as Residential Area except DDES 5 and DDES 6 which are classified as Non-residential Area. Annual dust fallout concentrations for each location are presented in Table 5-5. The monthly concentrations for the period are presented in Figure 5-2. A detailed description of the monthly dust fallout concentrations is included in Appendix A. The monthly dust fallout concentrations (Figure 5-2) shows that dust fallout concentrations exceeded the Residential Area limit of 600 mg/m 2 /day 361 times out of 1118 (32.3%) recorded dust fallout measurements at Residential Area monitoring locations. The Non-residential Area limit was exceeded 27 times out of 119 (22.7%) dust fallout measurements at Non-residential points. Rainfall does not appear to play a significant role in reducing dust fallout concentrations since numerous instances of dust fallout concentrations exceeding the Residential Area and Non-residential Area limits are noted during spring and summer when the majority of rainfall for the area is received.

41 SRK Consulting: : Dorstfontein East AQIA Page 40 Table 5-4: Dust monitoring locations Monitoring Point Location Designated Area DDES '54,6" 29 35'80,4" Residential Area DDES '99,1" 29 33'98,5" Residential Area DDES '02,1" 29 28'47,9" Residential Area DDES '88,0" 29 35'31,1" Residential Area DDES '04,2" 29 35'13,5" Non-residential Area DDES '04,9" 29 35'53,3" Non-residential Area DDES '46,1" 29 36'26,8" Residential Area DDES ,5" ,5" Residential Area DDES ,6" 29 21,37.5" Residential Area DDES ,7" ,2" Residential Area DDES ,0" ,4" Residential Area DDES ,29, '21.8 Residential Area DDES ,93, ,60.0 Residential Area DDES ,60, ,34.2 Residential Area DDES ,59, ,05.3 Residential Area DRL '14,5" 29 34'20,2" Residential Area DRL '09,8" 29 35'04,1" Residential Area DDED 1N 26 20'05,9" 29 32'57,0" Residential Area DDED 1E 26 20'05,9" 29 32'57,0" Residential Area DDED 1S 26 20'05,9" 29 32'57,0" Residential Area DDED 1W 26 20'05,9" 29 32'57,0" Residential Area DDED 2N 26 21'49,0" 29 34'49,4" Residential Area DDED 2E 26 21'49,0" 29 34'49,4" Residential Area DDED 2S 26 21'49,0" 29 34'49,4" Residential Area DDED 2W 26 21'49,0" 29 34'49,4" Residential Area

42 SRK Consulting: : Dorstfontein East AQIA Page 41 Figure 5-1: Location of monitoring points

43 SRK Consulting: : Dorstfontein East AQIA Page 42 Table 5-5: Average monthly dust fallout concentrations for each year for the period March 2012 to February 2017 Monitoring Point DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES 12 NM DDES 13 NM NM NM DDES 14 NM NM NM DDES 15 NM NM NM DRL DRL DDED 1N DDED 1E DDED 1S DDED 1W DDED 2N DDED 2E DDED 2S DDED 2W Residential Area Non-residential Area

44 SRK Consulting: : Dorstfontein East AQIA Page Dust fallout data for East Dorstfontein for the period March 2012 to February Dust fallout (mg/m 2 /day) Month DDES 1 DDES 2 DDES 3 DDES 4 DDES 5 DDES 6 DDES 7 DDES 8 DDES 9 DDES 10 DDES 11 DDES 12 DDES 13 DDES 14 DDES 15 DRL 1 DRL 2 DDED 1N DDED 1E DDED 1S DDED 1W DDED 2N DDED 2E DDED 2S DDED 2W Residential Area Non-residential Area Figure 5-2: Monthly dust fallout concentrations at East Dorstfontein for the period March 2012 to February 2017

45 SRK Consulting: : Dorstfontein East AQIA Page 44 6 Dispersion Modelling Methodology Dispersion models compute ambient concentrations and deposition levels as a function of source configuration, emission strengths and meteorological characteristics thus, providing useful tools to ascertain the spatial and temporal patterns in the ground level concentrations and deposition arising from the emissions of various sources. The US-EPA AERMOD modelling software package was used to model PM 10, PM 2.5 and dust fallout emissions, respectively, resulting from the proposed development. 6.1 Model Used Dispersion models compute ambient concentrations and deposition levels as a function of source configuration, emission strengths and meteorological characteristics, thus providing useful tools to ascertain the spatial and temporal patterns in the ground level concentrations and deposition arising from the emissions of various sources that are modelled. The US EPA AERMOD modelling software package (AERMOD View 8.2) was used to model PM 10 PM 2.5 and dust fallout emissions, respectively, resulting from the proposed development. The AERMOD model is included in a suite of models used by the US-EPA for regulatory purposes. The models have also been approved for use in South Africa. 6.2 Quantification of Emissions Data used in the modelling of emissions should ideally come from source-specific emission tests or continuous emission monitors. However, for a few sources such the haul roads and open pits measured data were not available for this project. Therefore, emission factors were used as they are the best surrogate measure available for quantifying emissions. In order to calculate emission rates, information regarding the proposed project activities were used determine emission factors. In this case, the US EPA emission factors in its AP 42 document, Compilation of Air Pollution Emission Factors were used. The AP 42 emission factors are the most widely used in the field of air pollution and are regularly subjected to revision and review as more effective modelling techniques are developed. Empirically derived, predictive emission factor equations are available for dust entrained by vehicles from both paved and unpaved roads, material handling operations and wind erosion. The impact of fugitive dust on air quality depends largely on the extent of the drift potential of the particles. The drift potential of particles depends in turn on the initial height of the emission, the terminal settling velocity of the particle (which in turn is a function of the particle diameter) and the degree of atmospheric turbulence. Larger dust particles tend to settle out near the source, creating a local nuisance problem, whereas finer particles are more likely to be dispersed over greater distances since their settling rate is retarded by atmospheric turbulence. Predictive emission factor equations allow for the estimation of emissions for particular particle size ranges, i.e. PM 10, PM 2.5 and dust fallout (US EPA, 2005). Appendix B includes a detailed description how the emissions used for the dispersion model were quantified. 6.3 Specify Model Settings Land Use Classification For most applications, the Code of Practice recommends the Land Use Procedure as sufficient for determining the urban/rural status of a modelling domain. The alternative approach on urban/rural classification using the Population Density Procedure is not encouraged as the approach can be rather subjective.

46 SRK Consulting: : Dorstfontein East AQIA Page 45 The classification of a site as urban or rural is based on the Auer method specified in the US EPA guideline on air dispersion models. From the Auer's method, areas typically defined as Rural include: Residences with grass lawns and trees. Large estates. Metropolitan parks and golf courses. Agricultural areas. Undeveloped land. Water surfaces. An area is defined as Urban if it has less than 35 percent vegetation coverage or the area falls into one of the use types described in Table 6-1. Table 6-1: Land types, -use, structures and vegetation cover Type Use and Structures Vegetation I1 Heavy Industrial Less than 5 % I2 Light/moderate industrial Less than 5 % C1 Commercial Less than 15 % R2 Dense single / multi-family Less than 30 % R3 Multi-family, two-story Less than 35 % Note: Source DEA, Code of Practice for Air Dispersion Modelling The area within and surrounding the project area has the following land uses: urban, cultivated land, water surfaces and natural land Surface Roughness The surface roughness length is not a physical height but a theoretical one based on the wind profile. It is the height at which the mean horizontal wind speed approaches zero. For many modelling applications, the surface roughness length is typically about an order of magnitude (factor of 10 or so) smaller than the average heights of the roughness elements. The surface roughness length can be adequately estimated from the land use categories as a function of the season. The average surface roughness included into the model for each land use is presented in Table 6-2. Table 6-2: Surface roughness Land use Surface roughness Urban 1 Cultivation Water Natural Albedo Noon-time albedo is the fraction of the incoming solar radiation that is reflected from the ground when the sun is directly overhead. For practical purposes, the selection of a single value for noon-time albedo to process a complete year of meteorological data is desirable. If other conditions are used, the regulatory agency must review the proposed noon-time albedo values used to pre-process the meteorological data. The average albedo included into the model for each land use is presented in Table 6-3.

47 SRK Consulting: : Dorstfontein East AQIA Page 46 Table 6-3: Albedo Land use Albedo Urban Cultivation 0.28 Water 0.14 Natural Bowen Ration The daytime Bowen ratio, an indicator of surface moisture, is the ratio of sensible heat flux to latent heat flux and is used for determining planetary boundary layer parameters for convective conditions driven by the surface sensible heat flux. The presence of moisture at the Earth's surface alters the energy balance, which in turn alters the sensible heat flux and Monin-Obukhov length. Bowen ratio values vary depending on the surface wetness. The average moisture conditions must be used normally, and the regulatory agency should review any other proposed Bowen ratio values used to pre-process the meteorological data. The average Bowen ratio included into the model for each land use is presented in Table 6-4. Table 6-4: Bowen Ration Land use Bowen Ration Urban Cultivation 0.75 Water 0.45 Natural Description of Modelled Meteorological Data The AERMOD model requires hourly average meteorological data as input, including wind speed, wind direction, a measure of atmospheric turbulence, ambient air temperature and mixing height. In the absence of suitable long term hourly on-site data sets for all the input parameters required for modeling, a 3-year model ready meteorological dataset was purchased from Lakes. The Lakes data are based on synoptic weather data prepared using the MM5 model, which is used to forecast weather. For this model simulation exercise hourly meteorological data for the period 1 January 2013 to 31 December 2015 were used. 6.5 Discussion on Upper Air Data Utilised The upper air data is modelled data included in the package for the Modelled MM5 data. The data is in the prescribed TD Fixed length format which is ready for processing in AERMET. For this model simulation exercise upper air data for the period 1 January 2013 to 31 December 2015 was used. 6.6 Description of Receptor Grid The receptor grid for the dispersion of dust emissions from the proposed project, under various operating scenarios, covers an area of 400 km 2. The size of the Cartesian (X:Y) grid area is larger than the mining concession area so that when the model runs the extent of the concentration plume levels beyond the concession boundary can be determined. The modelling domain extends for 20 km from east to west and 20 km from north to south with the proposed project area in the center. The extent of the modelling also includes the closest sensitive receptors.

48 SRK Consulting: : Dorstfontein East AQIA Page 47 Discrete receptors i.e. villages and existing monitoring locations, were also added into the model. The discrete receptor locations are the center points of villages in close proximity to the mine area. The towns/villages/homesteads included in the model are presented in Table 6-5. Figure 6-1 presents the modelling domain and receptor grid together with the discrete receptors. Table 6-5: Receptors in close proximity to the site Town/Village X - co-ordinate Y- co-ordinate Kriel Thubelihle Boskrans Monitoring locations DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDED Description of Topographical Data The AERMOD model was set up in such a way as to incorporate the topography of the study area. Topography can influence meteorological parameters (such as temperature inversions, local wind circulations, etc.) that may worsen or improve air quality on a local scale. The Terrain Height Options section in the model was set to elevated terrain, which assumes that the receptors, such as towns/villages, may potentially be above the source base. Terrain data downloaded from WebGIS ( for the study area were processed using the Terrain processor function in AERMOD and imported into the model, yielding a 3D-model and because of this, all identified receptors were assigned heights when inserted into the model. A topographical map is presented in Figure 6-2. In a west to east direction across the site, there is undulating topography which varies in elevation between meters above mean sea level (mamsl). In a north to south direction across the site, higher elevations are observed to the south and decreases towards the north. Elevations across the north to south profile range from mamsl.

49 SRK Consulting: : Dorstfontein East AQIA Page 48 Figure 6-1: Receptor Grid:

50 SRK Consulting: : Dorstfontein East AQIA Page 49 Figure 6-2: Terrain in the project area

51 SRK Consulting: : Dorstfontein East AQIA Page 50 7 Dispersion Modelling Results Predicted 99 th percentile daily and annual average concentrations for PM 10, PM 2.5 and maximum predicted daily averages for dust fallout were simulated using the US EPA approved AERMOD model. Isopleth maps showing PM 10, PM 2.5 and dust fallout concentrations are presented in the sections that follow. It should be noted that only sources from the proposed development have been modelled, all sources from existing operations have been excluded. Modelled ambient ground level concentration isopleths represent interpolated values from the concentrations predicted by the AERMOD model for each of the receptor grid points. It should be noted that the plots reflecting daily averaging periods contain only the highest average predicted ground level concentrations over the entire period for which the simulations were undertaken. These high concentrations are usually associated with extreme meteorological conditions such as high winds and low level inversions. Prior to the interpretation of the results it is important to reiterate the degree of uncertainty associated with emissions inventory preparation and dispersion simulation in general. Various assumptions have been made regarding the nature, location and extent of certain activities during the operational phase of the project. Of importance, however, are the trends and the areas that are likely to be impacted. The 99 th percentile values were chosen for predicted PM 10 and PM 2.5 emission concentrations, as this is a normal statistical method to exclude anomalous predicted concentrations. This in turn identifies emission concentrations in a range that is most likely to be observed during the proposed project s operational phase and excludes outliers. 7.1 PM10 The PM 10 dispersion modelling results for the proposed operations are presented in Table 7-1. Predicted concentrations at the receptors identified in close proximity have also been included. Without management: The maximum predicted 99 th percentile 24-hr PM 10 concentration for the scenario with management is 646 µg/m³ (Figure 7-1). This concentration is observed to be c.481 meters to the northeast of the discard dump which is within the boundary of the mining area. The maximum predicted 99 th percentile PM 10 concentration is above the South African Standard of 75 µg/m³. The PM 10 concentrations decrease to below 75 µg/m³ with distance from the mine area. The annual predicted PM 10 concentration of 85 µg/m³ is above the NAAQS of 40 µg/m³ (Figure 7-2). Predicted 99 th percentile and annual PM 10 concentrations at the sensitive receptors i.e. Kriel, Thubelihle and Boskrans, are presented in Table 7-1. Predicted concentrations at Thubelihle and Boskrans exceed the 24 hour NAAQS of 75 µg/m³. It should be noted that the scenario without management represents a worst case scenario. With management: Based on information received from the client, it is expected that management measures will be implemented at the mine during the proposed project. The maximum predicted 99 th percentile PM 10 concentration with management measures in place decreases to 129 µg/m³. The maximum concentration also occurs to the northeast of the discard dump. The concentration is above the South African standard of 75 µg/m³, however concentrations above 75 µg/m³ only occur on-site. Predicted PM 10 concentrations decrease to levels well below the NAAQS. Predicted concentrations will continue to decrease to concentrations that are low to negligible beyond the boundaries of the mine. Predicted annual concentrations are also below the NAAQS, with concentrations at the receptors decreasing with management. Maximum predicted 99 th percentile concentrations at the sensitive receptors decrease to levels well below the 24 and annual NAAQS.

52 SRK Consulting: : Dorstfontein East AQIA Page 51 Table 7-1: Predicted PM 10 concentrations at selected receptor locations for the proposed project PM 10 Without management With Management 24 hour (99 th ) Annual 24 hour (99 th ) Annual Units µg/m 3 µg/m 3 µg/m 3 µg/m 3 Maximum predicted concentration Village Coordinates X Y 24 hour Annual 24 hour Annual Kriel Thubelihle Boskrans South African Standard

53 SRK Consulting: : Dorstfontein East AQIA Page 52 Figure 7-1: Maximum predicted 99 th percentile PM10 concentrations for the proposed project.

54 SRK Consulting: : Dorstfontein East AQIA Page 53 Figure 7-2: Maximum predicted Annual PM10 concentrations for the proposed project

55 SRK Consulting: : Dorstfontein East AQIA Page PM2.5 The PM 2.5 dispersion modelling results for the proposed project are presented in Table 7-2. Predicted concentrations at the receptors identified in close proximity have also been included. Without management: The maximum predicted 99 th percentile 24-hr PM 2.5 concentration for the scenario with management is 86.4 µg/m³ (Figure 7-3). This concentration is observed to be c.571 m to the southwest of the discard dump. The maximum predicted 99 th percentile PM 2.5 concentration is above the South African Standard of 40 µg/m³. The PM 2.5 concentrations decrease to levels below 40 µg/m³ within the mine footprint and the plume disperses adequately so as to not have an impact on any sensitive receptors. The annual predicted PM 2.5 concentration of 15.3 µg/m³ is below the NAAQS of 20 µg/m³. Predicted concentrations at the sensitive receptors are below the 24 hour NAAQS of 40 µg/m³. It should be noted that the scenario without management represents a worst case scenario. With management: The maximum predicted 99 th percentile PM 2.5 concentration with management is µg/m³. The maximum concentration occurs to the northeast of the proposed open pit. The concentration is below the South African standard of 40 µg/m³. The PM 2.5 footprint around the proposed activities decreases with management. Predicted concentrations will continue to decrease to concentrations that are low to negligible beyond the boundaries of the mine. Predicted annual concentrations are also below the NAAQS, with concentrations at the receptors decreasing with management. Table 7-2: Predicted PM 2.5 concentrations for the proposed project PM 2.5 Without management With Management 24 hour (99 th ) Annual 24 hour (99 th ) Annual Units µg/m 3 µg/m 3 µg/m 3 µg/m 3 Maximum predicted 99 th percentile Village X Coordinates Y 24 hour Annual 24 hour Annual Kriel Thubelihle Boskrans South African Standard

56 SRK Consulting: : Dorstfontein East AQIA Page 55 Figure 7-3: Maximum predicted 99 th percentile PM2.5 concentrations for the proposed project

57 SRK Consulting: : Dorstfontein East AQIA Page 56 Figure 7-4: Maximum predicted annual PM2.5 concentrations for the proposed project

58 SRK Consulting: : Dorstfontein East AQIA Page Dust Fallout The dust fallout dispersion modelling results for the proposed project are presented in Table 7-3. Predicted concentrations at the receptors identified in close proximity have also been included. Without management: The maximum predicted dust fallout concentration for the scenario with management is 651 mg/m 2 /day (Figure 7-3). The maximum predicted concentration occurs at monitoring point DDES05. The maximum predicted dust fallout concentration is below the National Dust Control Regulations (NDCR) Non-residential Area Standard of 1,200 mg/m 2 /day. The dust fallout concentrations decrease within the footprint of the road and the plume disperses adequately so as to not have an impact on any sensitive receptors. Predicted dust fallout concentrations at the sensitive receptors are below the NDCR Residential Area standard of 1,200 mg/m 2 /day at all receptors. At the towns of Kriel, Thubelihle and Boskrans, the predicted 24 hour dust fallout concentrations are well below the NCDR. With management: The maximum predicted dust fallout concentration with management is 136 mg/m 2 /day. The maximum concentration also occurs at monitoring point DDES05. The concentration is below the NDCR Non-residential area standard of 1,200 mg/m 2 /day. The dust fallout footprint around the road decreases with management. Predicted concentrations will continue to decrease to concentrations that are low to negligible beyond the boundaries of the mine. Predicted concentrations at the receptors decrease with management. Table 7-3: Predicted dust fallout concentrations Dust fallout Without management 24 hour 24 hour With management Units mg/m 2 /day mg/m 2 /day Maximum predicted concentration Village Coordinates X Y 24 hour 24 hour Kriel Thubelihle Boskrans DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES DDES

59 SRK Consulting: : Dorstfontein East AQIA Page 58 Dust fallout Without management With management 24 hour 24 hour Units mg/m 2 /day mg/m 2 /day DDES DDED NDCR Residential Area Limit NDCR Non-residential Area Limit 1,200 1,200

60 SRK Consulting: : Dorstfontein East AQIA Page 59 Figure 7-5: Maximum predicted 99 th percentile PM2.5 concentrations for the proposed project

61 SRK Consulting: : Dorstfontein East AQIA Page 60 8 Impact Assessment The impact assessment focuses on the proposed operational phases of the project. The change in air quality as a result of a change in ambient PM 10, PM 2.5 and dust fallout levels during the proposed operational phase of the mine were the only impacts that were assessed. Residents in the towns surrounding the mining areas are the potential sensitive receptors for this study. This includes the surrounding towns of Kriel, Thubelihle and Boskrans. The impact assessment methodology used for this study was provided by SRK. The significance of the impacts was assessed according to the following criteria: Severity magnitude or intensity. Duration (temporal influence) Temporal influence. Spatial scope Scale. 8.1 Methodology Table 8-1: Criteria for Assessing Significance of Impacts SEVERITY OF IMPACT (magnitude) RATING Insignificant / non-harmful 1 Small / potentially harmful 2 Significant / slightly harmful 3 Great / harmful 4 Disastrous / extremely harmful 5 SPATIAL SCOPE OF IMPACT (Extent) RATING Activity specific 1 Mine specific (within the mine boundary) 2 Local area 3 Regional 4 National 5 CONSEQUENCE DURATION OF IMPACT (temporal scale) RATING One day to one month 1 One month to one year 2 One year to ten years 3 Life of operation 4 Post closure / permanent 5 FREQUENCY OF ACTIVITY / DURATION OF RATING ASPECT Annually or less / low 1 6 monthly / temporary 2 Monthly / infrequent 3 Weekly / life of operation / regularly / likely 4 Daily / permanent / high 5 FREQUENCY OF IMPACT RATING Almost never / almost impossible 1 Very seldom / highly unlikely 2 Infrequent / unlikely / seldom 3 Often / regularly / likely / possible 4 Daily / highly likely / definitely 5 LIKELIHOOD (Probability)

62 SRK Consulting: : Dorstfontein East AQIA Page 61 Table 8-2: Interpretation of Impact Rating Consequence Likelihood High 76 to 150 Improve current management Medium High 40 to 75 Medium Low 26 to 39 Maintain current management Low 1 to 25 No management required SIGNIFICANCE = CONSEQUENCE x LIKELIHOOD Construction Impacts associated with dust (dust fallout, PM10 and PM2.5) no management measures The impact is considered to be of low significance before management and determined as follows: Severity: The severity of the impact is insignificant as the expected emissions of dust and expected contribution to ambient air quality during the construction phase for the proposed project should be below the NAAQS and NDCR. The impacts will be closer to construction activities and will decrease further away. Towns and surrounding farms in range of the construction activities will not be impacted as the expected contribution to levels of dust are expected to be low. Spatial: The impact will be activity specific as the dust concentrations will occur around the sources and decrease significantly away from the operational activities. Duration: Dust emissions are expected to impact from one month to a year and mainly during the dry season. The wet season may act as a natural mitigation measure, hence the duration of the impact is over a shorter period; however the effectiveness of this will need to be determined. Frequency of activity: The activity will occur for a set period as it only involves the construction phase of the project. Frequency of impact: The impact will be highly unlikely as dust generated may be of low concentrations. Impacts associated with dust (dust fallout, PM10 and PM2.5) with management measures The impact is considered to be of low significance after management and determined as follows: Severity: The severity of the impact is insignificant as the expected emissions of dust and expected contribution to ambient air quality during the construction phase for the proposed project with management will be below the NAAQS and NDCR. The impacts will be closer to construction activities and will decrease further away. Towns and surrounding farms in range of the construction activities will not be impacted as the expected contribution to levels of dust are expected to be low.

63 SRK Consulting: : Dorstfontein East AQIA Page 62 Spatial: The impact will be activity specific as the dust concentrations will occur around the sources and decrease significantly away from the operational activities. Duration: Dust emissions are expected to impact from one month to a year and mainly during the dry season. The wet season will act as a natural mitigation measure, hence the duration of the impact is over a shorter period. Frequency of activity: The activity will occur for a set period as it only involves the construction phase of the project. Frequency of impact: The impact will be highly unlikely because dust generated may be of low concentrations. Table 8-3: Impacts during the construction phase Impact Consequence Likelihood Significance Severity Spatial Duration Frequency Frequency Rating of Activity of Impact Impact before management Key management measures: Mitigation Rating Low Regular irrigation by water especially during windy conditions at the site, access road and construction material and debris with just enough moisture to keep the dust down without creating significant runoff. Should water not be available as a result of drought conditions then chemical suppressants need to be considered. Reduction of speed on unpaved roads to reduce the entrainment of dust into the atmosphere. During grading activities, any exposed earth should be watered if it is going to be exposed for long periods. If dust generating material such as soil, waste rock is hauled from the site, vehicles should be covered with a tarpaulin to reduce spillages. On windy days, or when fugitive dust is dispersed from the Site of Works, additional application of water to the affected areas should be applied. Impact after management 8.2 Operational phase Low Impacts associated with dust (dust fallout, PM10 and PM2.5) No management measures The impact is considered to be of medium-high significance before management and determined as follows: Severity: The severity of the impact is significant because the expected emissions of dust and expected contribution to ambient air quality during the operational phase are above the 24 hour NAAQS for PM 10 and PM 2.5, however below the NDCR for dust pollutants. The impacts are closer to operational activities and decrease further away. Towns and surrounding farms in range of the operational activities but will not be impacted as the expected contribution to levels of dust is not harmful. Spatial: The impact will be local area specific as the predicted dust concentrations are above the NAAQS for PM 10 beyond the fence line of the mine. Duration: PM 10 and PM 2.5 emissions are expected to impact from one month to a year and mainly during the dry season. The wet season will act as a natural mitigation measure, hence the duration of the impact is over a shorter period. Frequency of activity: The activity will occur during the life of operation of the mine as there will always be exposed areas and dust generating activities at the mine that will allow dust to be emitted into the atmosphere. Frequency of impact: The impact will be often because the predicted concentrations are significantly higher than the NAAQS.

64 SRK Consulting: : Dorstfontein East AQIA Page 63 Impacts associated with dust (dust fallout, PM10 and PM2.5) With management measures The impact is considered to be of low significance after management determined as follows: Severity: The severity of the impact is insignificant as the expected emissions of dust and expected contribution to ambient air quality during the operational phase for are low and below the 24 hour NAAQS and the NDCR for dust pollutants beyond the fenceline of the mine. The impacts are closer to operational activities and decrease further away. Towns and surrounding farms in range of the operational activities will not be impacted as the expected contribution to levels of dust is not harmful. Spatial: The impact will be activity specific as the predicted dust concentrations occur around the source but decrease significantly away from the operational activities. Duration: PM 10 and PM 2.5 emissions are expected to impact from one day to one month and mainly during the dry season. The wet season will act as a natural mitigation measure, hence the duration of the impact is over a shorter period. Frequency of activity: The activity will occur during the life of operation of the mine as there will always be exposed areas and dust generating activities at the mine that will allow dust to be emitted into the atmosphere. Frequency of impact: The impact will be almost impossible as the low levels that are currently observed are well below the NAAQS standards and NDCR. Table 8-4: Impacts during the operational phase Impact Consequence Likelihood Significance Severity Spatial Duration Frequency Frequency Rating of Activity of Impact Impact before management Key management measures: Mitigation Rating Mediumhigh When and where applicable, soil stockpiles that will not be used should be re-vegetated as soon as possible, or kept wet during windy periods. During the operational phases for the proposed project any bare ground surrounding the main operational area but within the boundaries of the facility must be covered with suitable vegetation that will be able to grow in the area. When fugitive dust can be observed leaving the area, additional dust suppression should be applied to the affected areas. If possible, additional dust monitoring equipment needs to be installed in order to effectively monitor dust related impacts from the proposed project area to the northwest and thereafter dust emissions can be managed better. In places of high vehicular traffic on unpaved roads, dust suppression measures on the roads should be implemented to reduce dust levels from the entrainment of dust. These measures will range from watering of roads, application of a chemical dust suppressant where watering is impractical, and/or paving of roads. Reduce the possibility of spillage from vehicles by ensuring all loads are covered, for example, with tarpaulin. Impact after management Low

65 SRK Consulting: : Dorstfontein East AQIA Page 64 9 Conclusions and Recommendations 9.1 Conclusions Based on the findings of this assessment the following were concluded: Based on the meteorological data from the on-site weather station, the climate in the project area is expected to be seasonal, with distinct warm months (September to February) and cool months (March to August). Rainfall is higher during the months from October to April and lower from May to September. The prevailing winds for this period are from the north, west-southwest and southeast, with lower occurrences from the east, east-southeast and south-southeast. The average wind speed measured for all hours is 4.47 m/s with maximum speeds less than 11.1 m/s and with calms (<0.5 m/s) of 1.72%. The monthly dust fallout concentrations shows that dust fallout concentration have exceeded the Residential Area limit of 600 mg/m 2 /day 361 times out of 1118 (32.3%) recorded dust fallout measurements. The Non-residential Area limit was exceeded 27 times out of 119 (22.7%) dust fallout measurements. The towns in the surrounding area i.e. Kriel, Tubelihle and Boskrans, have been identified as potential areas of impact, at the commencement of the study. The dispersion model was set up to predict PM 10, PM 2.5 and dust fallout concentrations from the proposed development at these towns and surrounding areas. Based on the nature of the activities and the dispersion modelling results, the receptors are unlikely to be impacted should management measures be implemented. With respect to dust (dust fallout, PM 10 and PM 2.5) concentrations, the predicted emissions resulting from the proposed operations were determined to be medium-high without management and low with management measures in place. Dust fallout concentrations will not exceed the NDCR limits at the closest sensitive receptors. Based on the findings of this assessment the proposed project will result in dust generation, however, the concentrations will be low and below the NAAQS (PM 10 and PM 2.5) and the NDCR (dust fallout) at the closest receptors, should management measures be implemented. Receptors in close proximity will not be impacted by the proposed project as impacts are predicted to be low beyond the boundary of the mine. In addition the peak concentrations are predicted to occur closer to the new activities. 9.2 Recommendations Based on the findings of this assessment the following are recommended: Maintain the current monitoring network and where necessary make minor adjustments to accommodate the installation of new infrastructure from the proposed project. In instances where activities change the monitoring network may be adjusted accordingly. When the proposed project is operational an annual monitoring program should be maintained to determine whether the proposed project is having an impact on the surrounding environment with respect to dust fallout. A continuous PM 10 and PM 2.5 monitor should be installed at the mine or if possible at sensitive receptors in close proximity to the mine. Conduct periodic independent audits of monitoring systems and the implementation of management plans to ensure that the system is maintained and that suitable data is obtained for decision making. When fugitive dust can be observed leaving the area additional dust suppression should be applied to the affected areas. The following mitigation measures are proposed and may be considered, if practicable, for the new activities: Regular irrigation by water, especially during windy conditions at the site, of the access road and construction material and debris with just enough moisture to keep the dust down without creating runoff. Should water not be available then chemical suppressants need to be considered. Reduction of speed on unpaved roads to reduce the entrainment of dust into the atmosphere.

66 SRK Consulting: : Dorstfontein East AQIA Page 65 During grading activities, any exposed earth should be watered if it is going to be exposed for long periods, especially during windy conditions. On windy days, or when fugitive dust is dispersed from the Site of Works, additional application of water to the affected areas should be implemented. When and where applicable, soil stockpiles that will not be used should be re-vegetated as soon as possible, or kept wet during windy periods. During the operational phases for the proposed project any bare ground surrounding the main operational area but within the boundaries of the facility must be covered with suitable vegetation that will be able to grow in the area. When fugitive dust can be observed dispersing from the site, additional dust suppression should be applied to the affected areas. If possible, additional dust monitoring equipment needs to be installed in order to effectively monitor dust related impacts from the project area and thereafter dust emissions can be managed better. In places of high vehicular traffic on unpaved roads, dust suppression measures on the roads may be implemented to reduce dust levels from the entrainment of dust. These measures will range from watering of roads, application of a chemical dust suppressant where watering is impractical, and/or paving of roads. Reduce the possibility of spillage from vehicles by ensuring all loads are covered, for example, with tarpaulin. Prepared by D. Naidoo (Pr. Sci. Nat) Senior Scientist Reviewed by V.S. Reddy (Pr. Sci. Nat) Partner All data used as source material plus the text, tables, figures, and attachments of this document have been reviewed and prepared in accordance with generally accepted professional engineering and environmental practices.